ORIGINAL ARTICLE

Early Enzymatic Burn Debridement: Results of the

DETECT Multicenter Randomized Controlled Trial

Downloaded from [Link] by guest on 11 October 2023

Yaron Shoham, MD1, ; Lior Rosenberg, MD1; William Hickerson, MD2,3, ; Jeremy Goverman, MD4;
Narayan Iyer, PhD5; Julio Barrera-Oro, PhD, MD5; Bretislav Lipovy, MD6, ; Stan Monstrey, MD7;
Sigrid Blome-Eberwein, MD8; Lucy A Wibbenmeyer, MD9; Martin Scharpenberg, PhD10, ;
Adam J. Singer, MD11,*, ; for the DETECT Investigators

Since 1970 surgeons have managed deep burns by surgical debridement and autografting. We tested the
hypothesis that enzymatic debridement with NexoBrid would remove the eschar reducing surgery and achieve
comparable long-term outcomes as standard of care (SOC). In this Phase 3 trial, we randomly assigned adults
with deep burns (covering 3–30% of total body surface area [TBSA]) to NexoBrid, surgical or nonsurgical SOC,
or placebo Gel Vehicle (GV) in a [Link] ratio. The primary endpoint was complete eschar removal (ER) at the
end of the debridement phase. Secondary outcomes were need for surgery, time to complete ER, and blood
loss. Safety endpoints included wound closure and 12 and 24-months cosmesis on the Modified Vancouver
Scar Scale. Patients were randomized to NexoBrid (n = 75), SOC (n = 75), and GV (n = 25). Complete ER
was higher in the NexoBrid versus the GV group (93% vs 4%; P < .001). Surgical excision was lower in the
NexoBrid vs the SOC group (4% vs 72%; P < .001). Median time to ER was 1.2 and 3.9 days for the NexoBrid
and SOC respectively (P < .001). ER blood loss was lower in the NexoBrid than the SOC group (14 ± 512
mL vs 814 ± 1020 mL, respectively; P < .0001). MVSS scores at 12 and 24 months were noninferior in the
NexoBrid versus SOC groups (3.7 ± 2.1 vs 5.0 ± 3.1 for the 12 months and 3.04 ± 2.2 vs 3.30 ± 2.76 for the
24 months). NexoBrid resulted in early complete ER in >90% of burn patients, reduced surgery and blood
loss. NexoBrid was safe and well tolerated without deleterious effects on wound closure and scarring.

Key words: burns; enzymatic debridement; eschar; surgery; excision; grafting.

Department of Plastic Surgery and Burn Unit, Soroka University Medical

1
(Funding acquisition [Equal]; Methodology [Equal]; Project administration
Center, Faculty of Health Sciences, Ben-Gurion University of the Negev, Beer- [Equal]; Writing – review & editing [Equal]) Bretislav Lipovy, MD
Sheba 8400711, Israel; 2Department of Plastic Surgery, College of Medicine, (Investigation [Equal]; Writing – review & editing [Equal]), Stan Monstrey
University of Tennessee Health Science Center, Memphis, TN 38163, USA; (Investigation [Equal]; Writing – review & editing [Equal]), Sigrid Blome-
3
Department of Medicine, Firefighters Regional Burn Center, Regional Eberwein (Investigation [Equal]; Writing – review & editing [Equal]), Lucy
One Health, Memphis, TN 38163, USA; 4Department of Surgery, Sumner Wibbenmeyer (Investigation [Equal]; Writing – review & editing [Equal]),
Redstone Burn Center, Massachusetts General Hospital, Harvard Medical Martin Scharpenberg (Investigation [Equal]; Writing – review & editing
School, Boston, MA 02114, USA; 5Burn and Blast Medical Countermeasures [Equal]), and Adam J Singer (Investigation [Equal]; Writing – original
Program, Division of Chemical, Biological, Radiological/Nuclear draft [Lead]; Writing – review & editing [Equal])
Countermeasures (CBRN), Biomedical Advanced Research and Development Funding: NexoBrid development has been supported in part with federal
Authority (BARDA), Administration for Preparedness and Response (ASPR) funding from US Biomedical Advanced Research and Development Authority
20201, HHS; 6Department of Burns and Plastic Surgery, University Hospital (BARDA), Administration for Strategic Preparedness and Response (ASPR),
Brno, Faculty of Medicine, Masaryk University, Brno 60300, Czech Republic; within the US Department of Health and Human Services (HHS), under
7
Department of Plastic and Reconstructive Surgery and Burn Center, ongoing USG Contract number HHSO100201500035C. Contract number
University Hospital of Ghent, Ghent 9000, Belgium; 8Lehigh Valley Health HHSO100201500035C provided funding and technical support for the
Network, Allentown, PA 18102, USA; 9Department of Surgery, Carver pivotal US Phase 3 clinical study (DETECT) and the marketing approval
College of Medicine, University of Iowa Health Care, Iowa City, IA 52242, registration process for NexoBrid.
USA; 10Universität Bremen, Kompetenzzentrum für Klinische Studien Conflict of Interest: Y.S. is a consultant for MediWound and Vericel and
Bremen, Bremen 28359, Germany; 11Department of Emergency Medicine, owns stock in MediWound. L.R. is a consultant for MediWound and owns
Renaissance School of Medicine, Stony Brook University, Stony Brook, NY stock in MediWound. S.B.E. owns stock in Biomed Sciences. J.G. owns stock
11794, USA in MediWound and is a consultant and speaker for Vericel. W.H. received
[Link] number: NCT02148705. consulting fees from Access Pro Medium-C MO Burns and SiOx-Medical.
Author Contributions: Yaron Shoham (Investigation [Equal]; Writing – A.J.S. is a speaker for AstraZeneca, Janssen and Abbott and is a consultant for
original draft [Equal]; Writing – review & editing [Equal]), Lior Rosenberg AstraZeneca and Abbott. N.I., J.B.O., M.S., S.M. have no COI to disclose.
(Conceptualization [Lead]; Formal analysis [Equal]; Funding acquisition
*
Address correspondence to A.J.S. (email: [Link]@[Link])
[Lead]; Investigation [Equal]; Supervision [Equal]; Writing – original © The Author(s) 2023. Published by Oxford University Press on behalf of the
draft [Equal]; Writing – review & editing [Equal]), William Hickerson American Burn Association. All rights reserved. For permissions, please e-mail:
(Investigation [Equal]; Writing – review & editing [Equal]), Jeremy [Link]@[Link].
Goverman (Investigation [Equal]; Writing – review & editing [Equal]),
Narayan Iyer (Funding acquisition [Equal]; Project administration [Equal]; [Link]
Supervision [Equal]; Writing – review & editing [Equal]), Julio Barrera-Oro

1
 Journal of Burn Care & Research
2  Shoham et al XXXX/XXXX 2023

INTRODUCTION An independent data safety monitoring board (DSMB),

consisting of two experienced burn specialists and a biostat-
It is estimated that nearly 500 000 patients in the United istician, periodically convened in accordance with enroll-
States seek medical attention for burn injuries annually with ment rate to assess safety data.
about 40 000 requiring acute inpatient hospitalization.1 These
injuries lead to >3000 deaths in the United States alone.1–3 Patient selection and randomization
Improvements in resuscitation have led to reductions in mor-
Adult patients (ages 18 years and older) suffering from deep par-
tality.4 However, management of the burn wound itself re-
tial or full thickness burns (caused by flame, scald, or contact)
mains challenging.5
involving between 3% and 30% of their TBSA were eligible for
The current standard of care (SOC) for deep burns is removal

Downloaded from [Link] by guest on 11 October 2023

the study. Only patients who could be consented within 84 h of
of the eschar (debridement),6,7 predominantly by surgical exci-
injury were included. Study eligibility required that each patient
sion followed by skin grafting.8 While reducing mortality and
have a target wound involving any area of the body except for
scarring, surgical excision is traumatic and requires specialized
the face and perineum, which was deep partial or full thickness
personnel and facilities. Early enzymatic and selective debride-
and involved at least 0.5% TBSA. Exclusion criteria included
ment of the eschar is an alternative, nonsurgical modality that
patients with circumferential burns of the limbs, infected burns,
may, in many cases, obviate the need for surgical excision with
inhalation injury, pregnancy, and a major comorbidity.
its inherent drawbacks and complications.9,10 Anacaulase-bcdb
(NexoBrid, MediWound Ltd, Yavne, Israel) has been developed
Study design and treatments
and shown to reduce the overall need for and extent of surgery,
while reducing blood loss and achieving long-term functional and The DETECT study (NCT02148705; EudraCT 2014-
cosmetic outcomes comparable to those with surgical excisional 001672-55) was a Phase 3, randomized, controlled, assessor
debridement.9–14 While NexoBrid is approved for use in Europe blinded (2 endpoints), 3 arm, multicenter, international study
and other regions outside of the United States, this study was designed to evaluate NexoBrid treatment compared to GV
required for US Food and Drug Administration approval. (placebo control) and SOC (nonsurgical and surgical). The
In addition, the study offered the potential for approval of a study was conducted from May 2015 (first patient enrolled)
nonsurgical alternative for eschar removal (ER), with advantages to September 2019 (last patient completed).
in routine burn care and in burn mass casualty incidents, and An overview of the DETECT study design is shown in
funding was provided by the Biomedical Advanced Research and Figure 1. NexoBrid is a one- or two-time, 4-hour topical
Development Authority (BARDA) within the Administration application with a short systemic exposure; therefore, pri-
for Strategic Preparedness and Response (ASPR) in the US mary efficacy and safety assessments were performed in
Department of Health and Human Services (HHS). the acute phase which was defined as up to 3 months post
We conducted the Phase 3 DEbride and proTECT wound closure. In addition, as prespecified in the protocol,
(DETECT) trial to assess the efficacy and safety of enzy- safety data were collected in the DETECT study in both
matic debridement (ER) with NexoBrid when compared with the acute phase and in longer-term follow-up with a cutoff
placebo (Gel Vehicle [GV]), as well as reduction in surgical of 12 and 24 months after complete wound closure of all
burden and blood loss compared with SOC in adults with treated wounds.
deep burns. Prior to treatment, eligible patients were randomly assigned
in a [Link] ratio to receive NexoBrid, SOC, or placebo (GV).
Randomization was done in random and permuted blocks
METHODS stratified by trial center and burn size by means of a GCP elec-
tronic data capture web-based service.
Trial oversight
The study was designed and initially funded by MediWound
Treatments
Ltd. (Yavne, Israel). Subsequent funding and oversight
were provided by the Biomedical Advanced Research and An overview of the 3 treatment arms is depicted in Figure 2.
Development Authority (BARDA) within the Administration All patients who met eligibility criteria were to receive ER
for Strategic Preparedness and Response (ASPR) in the US treatment per the randomized treatment arm. Wound depth
Department of Health and Human Services (HHS). was assessed by clinical evaluation.15 Patients in all treatment
arms (NexoBrid, SOC, and GV) were treated in a similar way
except for the ER stage. Prior to initiation of ER, patients
Ethical considerations were medicated with appropriate analgesia and underwent
The study was conducted according to Good Clinical wound cleansing and dressing of all wounds with antibacterial
Practice (GCP) guidelines and principles of the Declaration solutions. Subsequently, patients underwent the ER process as
of Helsinki. Twenty-nine centers in 8 countries (United per treatment assignment (NexoBrid, SOC, or GV).
States, Belgium, Czech Republic, Germany, Romania, Israel,
Italy, and Georgia) enrolled and randomized patients into
the study. All study sites had written approval from their NexoBrid treatment:
Institutional Review Board (IRB)/Independent Ethics The overlying necrotic keratin layer (ie, the blisters) was
Committee (IEC) and local Competent Authority (as re- removed and the burn was soaked in an antibacterial solu-
quired locally), and all patients or their designees provided tion for at least 2 h. In patients assigned to NexoBrid the
written informed consent before participating in the study. enzymatic agent was applied at a dose of 2-gram sterile
Journal of Burn Care & Research
Volume XX, Number XX Shoham et al  3

Downloaded from [Link] by guest on 11 October 2023

Figure 1. Overall Study Design and Key Endpoints for the DETECT (MW2010-03-02) Study

Figure 2. Summary of Treatment Interventions (NXB = NexoBrid, SOC = Standard of Care)

 Journal of Burn Care & Research
4  Shoham et al XXXX/XXXX 2023

powder mixed with 20-gram sterile GV per 1% adult TBSA information on all surgical procedures including blood loss
burn. A barrier of petrolatum gel was applied adjacent to the and blood transfusions, percentage area of wound grafted,
burn edges. The wound was then covered with an occlusive graft take, size of donor sites, and need for scar modulation.
dressing in order to contain the enzymatic agent for 4 h. After
4 h, the dressings were removed and the enzymatic agent to-
gether with dissolved eschar was wiped with a wooden tongue Study endpoints/outcomes
depressor. The wounds were soaked in an antibacterial solu- An overview of the prespecified primary and secondary effi-
tion for an additional 2 h and then cleaned prior to assessment cacy endpoints and key safety outcomes is provided in Table
of ER. The amount of NexoBrid applied at any one session 1. The primary efficacy outcome was complete ER in the

Downloaded from [Link] by guest on 11 October 2023

was limited to 15% TBSA. Patients with burns greater than NexoBrid compared to GV (placebo) arm. The ER assess-
15% TBSA had the NexoBrid applied in two separate ses- ment was performed by a trained health professional who
sions. According to the study protocol, if ER was incomplete, was not involved in the treatment of the patient or wound, as
NexoBrid could be reapplied one additional time. the treating physician could distinguish between treatments
administered to each patient. For the topical arms (NexoBrid
and GV), ER assessment was performed immediately fol-
GV (placebo control): lowing removal of the soaking dressing (6 h after start of
In patients assigned to the placebo (GV), all study interventions first and second treatment and after any additional proce-
were as for NexoBrid except that only the topical gel, at a dose dure until complete ER). The dressing was soaked with anti-
of 20 gram per 1% TBSA burns, was applied (without the ac- bacterial solution, for example 3–5% Sulfamylon, 0.05–0.5%
tive enzyme powder). chlorhexidine, Dakin’s solution, hypertonic 5–10% saline so-
lution, or 0.9% saline, applied to the wound and left in place.
The assessment included wound depth assessment and clin-
SOC:
ical assessment of the extent of ER. In all 3 treatment arms,
Patients in the SOC arm may have been treated with a com- ER was considered complete when more than 95% of the
bination of surgical (eg, tangential excision, fascial exci- eschar was removed, as per the American Burn Association
sion, hydrosurgery, or dermabrasion) and nonsurgical (eg, consensus guidelines.7 Secondary efficacy outcomes in-
collagenase ointment, antimicrobial solutions, or silver cluded comparisons between NexoBrid and SOC in the need
dressings) ER procedures according to the investigator’s judg- for surgical excision, time to ER, and estimated blood loss.21
ment. If nonsurgical SOC treatment did not result in com- Safety endpoints included time to wound closure, and
plete ER, surgical SOC treatment may have been employed as long-term assessments of scar appearance and function.
a rescue procedure, according to the investigator’s judgment. Wounds were considered closed when fully re-epithelialized
The procedures were repeated as needed until complete ER. without any drainage or need for outer dressings and con-
In case of failure of debridement in patients treated by firmed at least two weeks later. Scar appearance was assessed
NexoBrid or the placebo GV a rescue SOC treatment could using the Modified Vancouver Scar Scale22 ranging from
be used at the discretion of the treating physician. 0 to 15 from best to worst. All wound assessments were
performed by observers blinded to all treatment assignments.
Wound closure Additional safety assessments included pain (VAS and ad-
verse events), level of sedation, and adverse events.
Following ER, the strategy used for wound bed closure was
at the discretion of the burn surgeon. If enough viable dermis
remained, the wound was allowed to re-epithelialize spon- Statistical analysis
taneously. In patients with inadequate viable dermis to sup- The study was powered at 90% to detect a difference between
port spontaneous wound re-epithelialization, wound closure groups for the primary and secondary endpoints based on
was achieved with autografting. Partial autografting of areas data obtained in earlier studies.23 The total sample size was
of deeper wounds or wounds with delayed epithelialization determined to be 121 patients (65 NexoBrid, 13 GV, and 43
was done according to the burn surgeon’s clinical judgment. SOC); however, the enrolled patient number was increased
Patients were then followed up for up to two years. to a total of 175 to provide adequate information on safety
outcomes.
All statistical analysis was predefined in a Statistical Analysis
Data collection Plan. Demographics and relevant baseline information are
We collected the data using standardized, structured data presented and summarized with appropriate descriptive sta-
collection forms that included medical history, physical ex- tistics. Chi-square tests (or, in case of small, estimated cell
amination, pain levels (using a 100 mm unhatched visual counts, Fisher’s exact test) for categorical variables and one-
analog scale [VAS] from 0 [none] to 100 [worst]), wound way analysis of variance for continuous variables were used
photographs, wound cultures, and central laboratory tests. to assess the comparability of the baseline characteristics be-
Burn depth assessments were performed by clinical evaluation tween the treatment arms. If any of the baseline characteristics
by experienced burn surgeons both before and after ER.16,17 were found to be significantly different between the treat-
The %TBSA of the burn was assessed by a burn surgeon using ment arms, then the factor was included as an extra adjusting
Lund and Browder charts18 or Wallace’s Rule of Nine,19 or for covariate in the supportive analysis models. Burn center was a
small burns: the patient’s own palm (including the fingers), covariate in secondary and sensitivity analyses of the primary
which was estimated as 1% TBSA,.20 We also collected detailed and secondary endpoints.
Journal of Burn Care & Research
Volume XX, Number XX Shoham et al  5

Table 1. Primary and Secondary Efficacy Endpoints and Key Safety

Primary efficacy endpoint
Complete Eschar Removal in the NexoBrid vs GV (placebo); The main analysis was based on the binary variable (yes/no): “has com-
topical treatment arms ­assessor blinded to treatment plete eschar removal been achieved in all TWs.”
Secondary efficacy endpoints
Time to complete eschar removal NexoBrid vs SOC Time (days) when complete eschar removal was achieved for each pa-
(days) tient from the time of randomization.
Reduction in surgical needs NexoBrid vs SOC Incidence of surgical eschar removal (tangential/minor/avulsion/
Versajet, and/or dermabrasion excision) in the NexoBrid compared

Downloaded from [Link] by guest on 11 October 2023

with SOC arm.
Amount of blood loss during NexoBrid vs SOC Actual blood loss (ABL), changes in hemoglobin during the eschar re-
­eschar removal process moval procedures, and units of blood transfused.
Safety endpoints and assessments
Wound Closure NexoBrid vs SOC Assessor Time to reach complete wound closure, assessed in days from randomi-
blinded to treatment zation. A noninferiority margin of 7 days was used in the analysis.
Cosmesis and Function (MVSS) NexoBrid, SOC, and GV As- Used to assess the quality of the wound closure scar at 1, 3, 6, 12, and
sessor blinded to treatment 24 months post wound closure
Level of Sedation NexoBrid, SOC, and GV Number and percentage of patients per each level of sedation and each es-
­(placebo) char removal procedure (in topical arms: first and second topical appli-
cation, surgical rescue procedures, and nonsurgical rescue procedures;
in SOC arms: surgical procedures and nonsurgical procedures)
Pain assessment NexoBrid, GV (placebo), Pain was assessed as a patient reported outcome using visual analog scale
and SOC [VAS], and as reported as an adverse event
Adverse Events NexoBrid, SOC, and GV Treatment emergent adverse events
(placebo)

GV = Gel Vehicle; MVSS = Modified Vancouver Scar Scale; SOC = Standard of Care.

For the primary and secondary endpoints all patients and quartiles. The treatment arms were compared using a Cox
randomized were included in the analysis in the group in which regression model.
they were randomized (full analysis set [FAS] =intention-to- The incidence of surgical excision was a binary yes/no
treat principle). For safety summaries, patients were included variable and the proportion of patients who needed excision
in the treatment arm in which they were treated. for ER were compared using logistic regression. The explan-
For the primary efficacy endpoint, the proportions of atory variables in the model included treatment and the fol-
patients who reached complete ER at the end of the topical lowing variables: overall TW depth (all TWs FT, mixed TWs,
agent soaking period were compared using logistic regres- and all TWs DPT), “Total % TBSA per patient,” and number
sion. The primary analysis was based on the binary variable of TWs (1, 2, and ≥3). The odds ratio of requiring surgery
(yes/no): “has complete eschar removal been achieved in all for NexoBrid versus SOC was estimated from the model, as
TWs” (as defined in study endpoints). The primary efficacy well as 95% CIs and the level of statistical significance.
comparison was between the NexoBrid and GV arms. The The measure of actual blood loss (ABL) was computed for
statistical test was based on Fisher’s exact test because of the each patient as described in the results section on blood loss,
small numbers expected in the GV treatment arm. The odds and the distribution in the NexoBrid arm was compared with
ratio of achieving complete ER for NexoBrid versus GV and that in the SOC arm. Means, standard deviations, medians,
its 95% confidence interval (CI) were estimated using exact and interquartile ranges were calculated. The normality of the
distribution methods. If assessment data of complete ER were data was tested on each treatment arm using the Shapiro-Wilk
missing, the patient was counted as having failed the endpoint test. If the normal distribution hypothesis was not rejected at
(ie, as not having achieved complete ER). the 0.5% significance level in either arm, then differences in
Time until complete ER was defined as the time from distribution between NexoBrid and SOC were tested using a
randomization date (in days) until complete ER had been t-test. If the normal distribution hypothesis was rejected either
achieved at a patient level (ie, for all TWs of an individual in the NexoBrid arm or in the SOC arm, then the differences
patient). For patients who did not reach complete ER, time in distribution between the treatment arms were tested using
was censored at the last nonmissing ER assessment (typically a Mann-Whitney test. Missing values were handled by the
the last debridement procedure). Kaplan-Meier curves were method of multiple imputation.
presented graphically to display the distribution of time to To preserve the overall significance level of each efficacy end-
complete ER under the 2 treatments (NexoBrid versus SOC). point, a hierarchical test procedure was implemented. Since
Median time to complete ER was estimated for each treat- highly statistically significant results were obtained for all primary
ment arm with a 95% CI. Additionally, time to complete ER and secondary endpoints, the testing procedure did not stop,
was analyzed descriptively with number of units, number of implying that all statistical tests of the primary and secondary
missing values, mean, standard deviation, min, max, median, endpoints primary analyses can be considered as confirmatory.
 Journal of Burn Care & Research
6  Shoham et al XXXX/XXXX 2023

The safety endpoint of time to wound closure was analyzed exploratory, and subgroup analyses (results not shown) con-
using the FAS. Time to reach complete wound closure (time sistently supported the primary analysis result. These con-
from randomization to confirmation of wound closure) was sistent results demonstrate that NexoBrid is a highly effective
compared between NexoBrid and SOC at a wound level enzymatic debriding agent.
using a method of survival analysis with clustered data that
is based on appropriate assumptions. “Clustered data” refers Time to complete ER (NexoBrid vs SOC)
to the multiple TWs that can occur in a patient. A non- The Kaplan-Meier estimates for time to complete ER (defined
inferiority (NI) margin was incorporated into the analysis that as time from the time of randomization until date of complete
represented a 7-day advantage to the SOC arm. After that, the ER) for the NexoBrid and SOC treatment arms in the FAS

Downloaded from [Link] by guest on 11 October 2023

proportional hazards assumption was checked in the same way (main analysis) are shown in Table 3. The estimated median
as in the analysis of the timely ER endpoint. time to complete ER was 1.0 and 3.8 days for the NexoBrid
The means and standard errors of the MVSS scores at 12 and SOC treatment arms, respectively (P < .0001).
and 24 months were estimated for each treatment arm. The
treatment arms were compared using a linear model with
Reduction in surgical needs (incidence of surgical excision)
MVSS scores as the dependent variable. A clinically mean-
(NexoBrid vs SOC)
ingful noninferiority margin was incorporated into the anal-
The proportion of patients who needed any surgical excision
ysis that represented 1.9 units or more advantage to the SOC
for ER was compared in the NexoBrid and SOC treatment
treatment arm for the MVSS analysis. No statistical analysis
arms using logistic regression. Surgical excision was required
was planned or performed for the pain assessment, level of
for ER in 72% of SOC and 4% of NexoBrid patients. The cal-
sedation, or for adverse events.
culated OR was 0.011 (P < .0001) (Table 4), meaning the
The data were analyzed with SAS (SAS Institute, Cary,
NexoBrid group had a 98.9% [(1 − 0.011) × 100] decrease
North Carolina), version 9.4.
in the odds of having surgical excision compared to the
odds of excision in the SOC group. The results of the sen-
RESULTS sitivity analyses (results not shown: per protocol [included
patients without major protocol violations], complete cases
Patient disposition and characteristics [only patients without missing values], and positive analyses
A patient disposition consort diagram is shown in Figure 3. [missing information counted as no surgical excision]) were
One-hundred and seventy-five patients were randomized in similar to the results of the main analysis (4.0%–4.05% inci-
the DETECT study: 75 to the NexoBrid arm, 75 to the SOC dence rates of surgical excisions and odds ratios of 0.010–
arm, and 25 to the GV arm (FAS equivalent to intention to 0.015 for patients in the NexoBrid treatment arm).
treat group). Of the 175 randomized patients, 169 received
the study treatment; 77 patients were treated with NexoBrid,
68 with SOC, and 24 with GV (safety set). ABL related to ER (NexoBrid vs SOC)
A similar percentage of patients across treatment arms The ABL formula takes into account the changes in hemo-
completed each phase of the study with most patients globin before and after the first debridement period as well as
completing the acute phase (84%–92% in all arms), and the volume of whole blood/PRBC transfused.
more than 75% completing the 12-month follow-up phase The ABL that occurred for each patient during ER was cal-
(76%–80% in all arms). At 24 months, the study completion culated as:
rates were, as expected, lower (57% in NexoBrid, 48% in the  
EBV ∗ H bbefore − H bafter 5
SOC, and 40% in the GV). The higher drop-out rate is not ABL =   + VWB + VPC
H bbefore + H bafter /2 3
uncommon among the burn population in long-term trials.24
Table 2 provides a summary of patient demographics and EBV = estimated blood volume assumed to be 70 cm3/kg,
burn characteristics. Patient characteristics and burn etiologies (Hbbefore − Hbafter) = Changes in hemoglobin (Hb) following
at baseline were similar across treatment groups. Most each ER procedure, VWB = Volume [mL] of whole blood
patients were White (79%–84%) males (60%–79%), with mean transfused, VPC = Volume [mL] of packed red blood cells
age of 41 years and body mass index of 27. The average time transfused, 5/3 = factor derived from Transfusion Medicine,
from injury to informed consent was from 33 to 38 h, and 4th Edition (Chapter 5); compensates for comparison of whole
the majority of patients had burns with fire/flame burn eti- blood and packed red blood cells.
ology (59%–84%). Mean %TBSA was similar across treatment There was significantly less ABL related to ER in the NexoBrid
groups (8.3%–9.0%);and the distribution of all target wounds compared with the SOC treatment arm. The mean ABL during
by %TBSA was similar across treatment groups for SPT, DPT, ER for patients in the NexoBrid arm was approximately 14 mL
and FT wounds. Representative images of a burn treated with compared with over 800 mL in the SOC arm (P-value < .0001).
NexoBrid are shown in Figure 4. Sensitivity analyses supported this primary analysis.

Efficacy results Safety results

Incidence of complete ER (NexoBrid vs GV) Wound closure and cosmesis and function were included in the
More than 93% of the patients treated with NexoBrid achieved DETECT study as safety endpoints. Adverse events and level of
complete debridement following 1 application of NexoBrid sedation are standard clinical trial summaries. Adverse events are
compared with 4% in the GV arm (P < .0001). All supportive, reported for the entire length of follow-up (up to 24 months).
Journal of Burn Care & Research
Volume XX, Number XX Shoham et al  7

Downloaded from [Link] by guest on 11 October 2023

Figure 3. Patient Disposition Consort Diagram. Details Regarding Randomization and Numbers of Patients Completing Each of the 3 Phases
of the Study (Acute [Primary Efficacy and Safety ≥3 Months Post Wound Closure], 12 Month [Long-term Safety Follow-up at 12 Months
Post Wound Closure], and 24 Month [Long-term Safety Follow-up at 12 Months Post Wound Closure]). For Efficacy Endpoints, All Patients
Randomized were Included in the Analysis in the Group in which They were Randomized (Full Analysis Set [FAS] = Intention-to-treat principle).
For Safety Summaries, Patients Were Included in the Treatment Arm in Which They were Treated (Safety Set)

Table 2. Patients’ and Wounds’ Baseline Characteristics

NexoBrid (N = 75) SOC (N = 75) Gel (N = 25)

Patients
 Age, mean (SD) 41.28 (15.03) 40.91 (15.16) 40.68 (17.30)
 Sex, male, n (%) 49 (65.33) 59 (78.67) 15 (60.00)
 Race, White n (%) 61 (81.3) 59 (78.7) 21 (84.0)
 BMI, mean (SD), kg/m2 27.64 (4.90) 26.56 (4.42) 27.02 (4.38)
Wounds
 Mean (SD) time from injury to informed consent, hours 37.62 (20.09) 37.98 (17.95) 33.35 (17.28)
 Etiology of injury, n (%)a
   Fire/flame 44 (58.7) 44 (58.7) 21 (84.0)
   Scald 22 (29.3) 18 (24.0) 2 (8.0)
   Contact 8 (10.7) 12 (16.0) 2 (8.0)
 Mean (SD) % TBSA per person all wounds 8.97 (5.18) 8.34 (4.24) 8.93 (3.63)
 Mean (SD) % TBSA per person all target wounds 6.28 (3.68) 5.91 (3.06) 6.53 (3.60)
 Wound distribution
  Mean (SD) %TBSA SPT 0.49 (0.85) 0.52 (0.90) 0.89 (1.33)
  Mean (SD) %TBSA DPT 2.24 (1.59) 2.20 (1.70) 2.07 (1.60)
  Mean (SD) %TBSA FT 0.95 (1.67) 0.71 (1.23) 0.84 (1.33)

BMI = body mass index; DPT = deep partial thickness; FT = full thickness; SD = standard deviation; SOC = standard of care, SPT = superficial partial thickness TBSA
= total body surface area.

Wound closure (NexoBrid vs SOC) debridement of burn wounds using NexoBrid had no delete-
The evaluation of wound closure as a safety endpoint was rious effect on the time to wound closure.
designed as a noninferiority test comparing time to wound Time to reach complete wound closure was compa-
closure between NexoBrid and SOC to ensure that enzymatic rable in the NexoBrid and SOC treatment arms. The
 Journal of Burn Care & Research
8  Shoham et al XXXX/XXXX 2023

Downloaded from [Link] by guest on 11 October 2023

Figure 4. Representative Images of a Deep Partial Thickness Burn Treated With NexoBrid. The Upper Images Show a Burn of the Thigh Before
(left) and After (Right) Enzymatic Debridement. The Lower Image Shows the Appearance of the Wound 1 Year after Injury

Table 3. Kaplan–Meier Estimates for Time to Complete Eschar Removal NexoBrid vs SOC
Treatment Median (days) Lower 95% confidence bound Upper 95% confidence bound

NexoBrid (75 patients) 1.0232 0.9827 1.0799

SOC (75 patients) 3.8279 1.9872 5.9849

FAS, full analysis set; SOC, standard of care.

P < .0001 (Generalized Wilcoxon-Gehan test adjusted for overall treated wound dept, TBSA group, center group, and number of treated wounds).

Kaplan-Meier estimated median time to complete wound clo- Cosmesis and function (NexoBrid, SOC, and GV)
sure for NexoBrid and SOC, was 27 and 28 days, respectively. The 12-month follow-up mean MVSS scores were lower
Statistical analysis established the noninferiority of NexoBrid (better) for the NexoBrid group (3.7 ± 2.1) than for the SOC
compared with SOC when incorporating a 7-day advantage (5.1 ± 3.1) and Gel groups (5.6 ± 3.0). A regression anal-
for the SOC group (P < .01). ysis showed that NexoBrid had a 1.4 MVSS point advantage
Journal of Burn Care & Research
Volume XX, Number XX Shoham et al  9

over SOC after adjustment for all other variables in the model

Upper 95% confidence bound

The main analysis of this endpoint used the FAS, which included 5 patients in the SOC group who were randomized but did not receive treatment. As detailed in the SAP, the missing values for these patients were included
(P-value = .0027). The 95% CI for this treatment excludes
the predefined noninferiority margin of 1.9 points, thus
establishing noninferiority of NexoBrid treatment compared
with SOC. Similar trends were observed in the 24-month

0.044
follow up mean MVSS scores. Results were slightly lower
(better) for the NexoBrod group (3.04 ± 2.2) than the SOC
(3.30 ± 2.76).

Downloaded from [Link] by guest on 11 October 2023

Level of sedation (acute phase), NexoBrid vs SOC (exploratory
endpoint)
The extent of analgesia and anesthesia use by the level of se-
dation in the NexoBrid, surgical SOC (nonsurgical SOC is
Lower 95% confidence bound

not expected to require sedation), and GV groups during the

Acute Phase is summarized in Table 5. Most patients treated
with NexoBrid required minimal or moderate sedation. The
use of general anesthesia was higher for patients treated
0.003

with SOC during surgical ER than for patients treated with

NexoBrid during first application (SOC = 87.5% [42/48]
compared with NexoBrid = 5% [4/77], respectively). Patients
treated with GV who required subsequent surgical excision
also required general anesthesia (12/13 [92%] patients).

Assessment of pain
Pain intensity was collected by VAS patient reported outcomes
2-sided P-value

scoring. Post first topical application, the VAS pain score was
<.0001

slightly higher in the NexoBrid (39.8) compared to the placebo

(33.8) group. The incidence of pain was also collected as an
adverse event. The incidence of pain was slightly less frequent
in the NexoBrid (6.5%) compared to the SOC (8.3%) group.

Adverse events (NexoBrid, SOC, and GV)

Wald χ2

Acute phase:
Test
Table 4. Incidence of Surgical Excision for Eschar Removal NexoBrid vs SOC

Treatment-emergent adverse events were observed across all

treatment groups. The most frequent adverse events (≥3% of
patients) in each treatment arm are shown in (Figure 5): A
total of 12 patients experienced serious adverse events during
Odds ratioa

the acute phase (6 NexoBrid, 4 SOC, and 3 GV patients).

0.011

All patients in all 3 treatment arms had serious events that

were mild to moderate with the exception of severe events
of sepsis and acute respiratory failure in patients treated with
as having had an excision. FAS, full analysis set; SOC, standard of care.

NexoBrid; acute respiratory distress syndrome, and septic

shock in patients treated with the SOC; and seizure and in-
Surgical excision incidence rate

fusion site thrombosis in patients treated with the GV. One

patient died due to a respiratory complication, assessed by the
Investigator and DSMB as not related to NexoBrid.
4.0%/72.0%

Adjusted for the other variables in the model.

Twelve-month follow-up:
As expected with a 1- or 2-time administration of a topical
treatment with short systemic exposure, there was a reduced
frequency of TEAEs past the first 3 months following wound
closure. Only 2 patients (both in the NexoBrid arm) experi-
enced an adverse event (folliculitis [mild] and pruritus [mod-
erate]) assessed by the investigator as related to study drug
in the 3- to 12-month period. One patient in the NexoBrid
arm died 8 months post wound closure period due to an un-
150 Patients:
75 NexoBrid

known cause following a second burn that underwent sur-

75 SOC

gical excision. The Investigator and DSMB assessed the death

as not related to study treatment.
n

a
 Journal of Burn Care & Research
10  Shoham et al XXXX/XXXX 2023

Table 5. Level of Sedation per First Topical Application and Surgical Excision (Safety Analysis Set)
Treatment/Procedure

NexoBrid/first NexoBrid/sur- SOC/sur- GV/first GV/surgical

Level of sedation application gical excision gical excision application excision

Overall, N 77 3 48 24 13
Minimal Sedation, n (%) 39 (51%) 0 4 (8%) 14 (58%) 1 (8%)
Moderate Sedation, n (%) 20 (25%) 0 2 (4%) 2 (8%) 0
Deep Sedation, n (%) 13 (17%) 0 0 1 (4%) 0

Downloaded from [Link] by guest on 11 October 2023

General Anesthesia, n (%) 4 (5%) 3 (100%) 42 (87.5%) 0 12 (92%)
Missing 1 (1.3%) 0 0 7 (29%) 0

Surgical excision and eschar removal are used interchangeably. Percentages are calculated as percentage within treatment arm and procedure. N = number of patients
within a treatment arm, n = number of observed patients within a treatment arm.

Figure 5. Treatment Emergent Adverse Events >3% Incidence in Any Arm (Acute Phase)

Twenty-Four-month follow-up: NexoBrid significantly reduced the need for surgery (number
There were no treatment related adverse events reported be- needed to treat 1.47; 95% CI, 1.28–1.85) as well as the as-
tween 12 and 24 months of follow-up. There were no deaths sociated blood loss. Complete ER was achieved at least 2
reported in the 24-month follow-up. days earlier with NexoBrid than with SOC. The time to com-
plete wound closure was similar in both NexoBrid and SOC
arms though excision and autografting is expected to close
DISCUSSION wounds faster than the slower process of epithelialization
over dermis. Long-term scar appearance in the NexoBrid
The DETECT study was conducted as part of the post ap- arm, as reflected by lower MVSS scores, was better at 1-year
proval commitments to the European Medicines Agency when compared to the SOC and GV arms, and was similar
(EMA) and to gain US FDA approval. As a result, its design at 24 months meeting the noninferiority test. The results of
and endpoints included the combined requirements of both the current study are in line with previous reports11–13,23,25–30
authorities. The results presented in this manuscript encom- while adding a placebo control arm and blinded assessment
pass the primary and secondary efficacy and main safety and of the primary outcome.
exploratory endpoints. Introduction of rapid enzymatic debridement as a
In this assessor-blinded (2 end points), controlled trial nonsurgical alternative for many deep burns is a significant
involving adult patients with deep burns covering 3–30% advance in burn care. Early ER and autologous skin grafting
TBSA, enzymatic debridement with NexoBrid was more ef- of deep burn wounds are considered one of the cornerstones
fective than its GV in removing the burn eschar, achieving of modern burn care as this reduces early complications and
complete ER in over 90% of patients. Compared with SOC, late sequelae, mainly scarring.8,31,32 Surgical debridement/
Journal of Burn Care & Research
Volume XX, Number XX Shoham et al  11

excision is currently recognized as the SOC for removal of 8. Orgill DP. Excision and skin grafting of thermal burns. N Engl J Med.
2009;360(9):893–901.
the eschar, however, this technique requires a high level of ex- 9. Edmondson SJ, Ali Jumabhoy I, Murray A. Time to start putting down
pertise in order to differentiate between viable and nonviable the knife: a systematic review of burns excision tools of randomised and
tissue. Surgical debridement is also associated with significant non-randomised trials. Burns. 2018;44(7):1721–1737.
10. Loo YL, Goh BKL, Jeffery S. An overview of the use of bromelain-based
blood and heat loss, and poor operator-dependent selectivity enzymatic debridement (Nexobrid®) in deep partial and full thickness
also results in viable tissues being sacrificed along with the es- burns: appraising the evidence. J Burn Care Res. 2018;39(6):932–938.
char.10 Consequentially, surgical debridement is often delayed 11. Rosenberg L, Krieger Y, Bogdanov-Berezovski A, Silberstein E,
Shoham Y, Singer AJ. A novel rapid and selective enzymatic debride-
until an accurate diagnosis of burn depth is reached confirming ment agent for burn wound management: a multi-center RCT. Burns.
the necessity for surgery, thus compromising the advantages of 2014;40(3):466–474.
12. Schulz A, Fuchs PC, Rothermundt I, et al. Enzymatic debridement

Downloaded from [Link] by guest on 11 October 2023

early ER. Alternatively, these disadvantages may lead clinicians of deeply burned faces: healing and early scarring based on tissue
to use traditional nonsurgical ER methods which are known to preservation compared to traditional surgical debridement. Burns.
be slow and much less effective than surgery, thus leading to 2017;43(6):1233–1243.
13. Schulz A, Shoham Y, Rosenberg L, et al. Enzymatic versus traditional
an increased complication rate and inferior long-term results. surgical debridement of severely burned hands: a comparison of selec-
An important advantage of rapid enzymatic debridement is tivity, efficacy, healing time, and three-month scar quality. J Burn Care
its selectivity exposing the underlying wound bed and allowing Res. 2017;38(4):e745–e755.
14. Bowers C, Randawa A, Sloan B, Anwar U, Phipps A, Muthayya P.
visual assessment of depth and viability of the dermis and its Enzymatic debridement in critically injured burn patients - our experi-
potential for spontaneous closure by epithelialization with less ence in the intensive care setting and during burn resuscitation. Burns.
2022;48(4):846–859.
need for skin grafting.33 This reduces or eliminates the mor- 15. Heimbach D, Engrav L, Grube B, Marvin J. Burn depth: a review. World
bidity associated with the skin graft donor wound (eg, pain and J Surg. 1992;16(1):10–15.
scarring). Earlier ER also reduces the release of inflammatory 16. Pham T, Gibran N, Heimbach D. Chapter 10 - evaluation of the burn
wound: management decisions. In: Herndon David N., ed. Total Burn
mediators and the risk of infection. Concerns that spontaneous Care. 3rd ed. W.B. Saunders; 2007:119–26.
re-epithelialization of the debrided wound bed would be delayed 17. Monstrey S, Hoeksema H, Verbelen J, Pirayesh A, Blondeel P.
(as opposed to immediate wound closure by autografting the sur- Assessment of burn depth and burn wound healing potential. Burns.
2008;34(6):761–769.
gically excised wound bed) and contribute to hypertrophic scar- 18. Lund CC, Browder NC. The estimation of areas of burns. Surg Gynecol
ring were unsubstantiated as wound closure and 1-year cosmesis Obstet. 1944;79(1):352–358.
19. Knaysi GA, Crikelair GF, Cosman B. The role of nines: its history and
were similar in burns treated with NexoBrid or the SOC. accuracy. Plast Reconstr Surg. 1968;41(6):560–563.
Strengths of our study include the relatively large sample 20. Rossiter ND, Chapman P, Haywood IA. How big is a hand? Burns.
size and long-term follow-up, as well as the use of multiple, 1996;22(3):230–231.
21. McCullough TC, Roth JV, Ginsberg PC, Harkaway Richard C. Estimated
well-validated outcomes. However, our study has several lim- blood loss underestimates calculated blood loss during radical retropubic
itations that need to be noted. Despite the relatively large prostatectomy. Urol Int. 2004;72(1):13–16.
number of burns, we could not fully account for several po- 22. Oliveira GV, Chinkes D, Mitchell C, Oliveras G, Hawkins HK, Herndon
DN. Objective assessment of burn scar vascularity, erythema, pliability,
tential confounding variables such as anatomical location, skin thickness, and planimetry. Dermatol Surg. 2005;31(1):48–58.
type, genetic predilection to scarring, and surgeon treatment 23. Rosenberg L, Shoham Y, Krieger Y, et al. Minimally invasive burn care: a
review of seven clinical studies of rapid and selective debridement using
preferences and experience. Finally, results in a well-controlled a bromelain-based debriding enzyme (Nexobrid®). Ann Burns Fire
study may not reflect those seen in real world settings. Disasters. 2015;28(4):264–274.
In conclusion, among adult patients with deep burns 24. Ohrtman EA, Zaninotto AL, Carvalho S, et al. Longitudinal clinical trial
recruitment and retention challenges in the burn population: lessons
involving 3–30% of their TBSA, NexoBrid was safe and effec- learned from a trial examining a novel intervention for chronic neuro-
tive at removing the burn eschar, reducing the need for sur- pathic symptoms. J Burn Care Res. 2019;40(6):792–795.
gery and blood loss while achieving 1-year cosmetic outcomes 25. Hirche C, Citterio A, Hoeksema H, et al. Eschar removal by bromelain
based enzymatic debridement (Nexobrid®) in burns: an European con-
at least as good as with surgery. The results of this study have sensus. Burns. 2017;43(8):1640–1653.
recently led to US FDA approval. 26. Hirche C, Kreken Almeland S, Dheansa B, et al. Eschar removal
by bromelain based enzymatic debridement (Nexobrid®) in burns:
European consensus guidelines update. Burns. 2020;46(4):782–796.
27. Di Lonardo A, Nardini V, De Rosa M, et al. Enzymatic escharolysis
REFERENCES with Nexobrid® on partial thickness burn wounds: pre- and post-
debridement histological assessment. Ann Burns Fire Disasters.
1. American Burn Association. Burn Incidence Fact Sheet. 2022. Accessed 2018;31(1):23–27.
November 20, 2022. [Link] 28. Schulz A, Ribitsch B, Fuchs PC, Lipensky A, Schiefer JL. Treatment of
burn-incidence-fact-sheet/. genital burn injuries: traditional procedures and new techniques. Adv
2. Crowe CS, Massenburg BB, Morrison SD, Naghavi M, Pham TN, Gibran Skin Wound Care. 2018;31(7):314–321.
NS. Trends of burn injury in the United States: 1990 to 2016. Ann Surg. 29. Schulz A, Perbix W, Shoham Y, et al. Our initial learning curve in the en-
2019;270(6):944–953. zymatic debridement of severely burned hands-management and pit falls
3. Peck MD. Epidemiology of burns throughout the world. Part I: distribu- of initial treatments and our development of a post debridement wound
tion and risk factors. Burns. 2011;37(7):1087–1100. treatment algorithm. Burns. 2017;43(2):326–336.
4. Latenser BA. Critical care of the burn patient: the first 48 hours. Crit 30. Ranno R, Vestita M, Maggio G, et al. Italian recommendations on enzy-
Care Med. 2009;37(10):2819–2826. matic debridement in burn surgery. Burns. 2021;47(2):408–416.
5. Brownson EG, Gibran NS. Evaluation of the burn wound: management 31. Sheridan RL. Comprehensive treatment of burns. Curr Probl Surg.
decisions. In: Total Burn Care. Elsevier; 2018:87–92. e2. 2001;38(9):657–756.
6. Gibran NS; Committee on Organization and Delivery of Burn Care, 32. Kagan RJ, Peck MD, Ahrenholz DH, et al. Surgical management of the
American Burn Association. Committee on Organization and Delivery of burn wound and use of skin substitutes: an expert panel white paper. J
Burn Care, American Burn Association. Practice Guidelines for burn care, Burn Care Res. 2013;34(2):e60–e79.
2006. J Burn Care Res. 2006;27(4):437–438. 33. Claes KEY, De Decker I, Monstrey S, et al. Helpful hints in de-
7. Gibran NS, Wiechman S, Meyer W, et al. Summary of the 2012 ABA Burn ciding what and when to operate after enzymatic debridement. Burns.
Quality Consensus Conference. J Burn Care Res. 2013;34(4):361–385. 2023;49(1):80–90.